Introduction

Idiopathic intracranial hypertension (IIH, also known as pseudotumor cerebri) is a disorder associated with increased intracranial pressure without clinical or radiological evidence of a space-occupying lesion and with normal cerebrospinal fluid (CSF) constituents [1–5]. The diagnosis is established according to the modified Dandy criteria [6]. The incidence of IIH in the general population is 0.9:100,000 [5]. It is well described in the adult population, and is usually associated with female gender, obesity, and child-bearing age. The natural history of IIH is unknown. In some cases, it is self-limited, while in others, the intracranial pressure remains elevated for many years, even if systemic and visual symptoms resolve. The effects of self-limited IIH on the visual system may be catastrophic; approximately 25% of patients may have significant visual impairment [7, 8]. The condition is relatively rare among young children, and only few studies have described it in the pediatric population. Some studies have suggested that the clinical profile of children with IIH may be different than that of their adult counterparts and that the precipitating factors may be considerably different as well [7, 9, 10]. In the current study, we assessed the visual outcome and the risk of recurrence among children aged 18 years and under who were diagnosed as having IIH.

Patients and materials

Patients and diagnostic procedures

We conducted a retrospective observational single-center study which was approved by the institutional ethics committee at the referral medical center. The children included in this study were identified from a database of IIH patients. The data was collected from the hospital IIH database. All of the consecutively enrolled children were younger than 18 years of age (range 20 months–17 years) when they were first diagnosed as having IIH according to the modified Dandy criteria [6]. All children underwent cerebral imaging and lumbar puncture to rule out other sources of intracranial pressure, and underwent comprehensive ophthalmic evaluations, including visual acuity and visual field tests. Information on height and weight was collected in order to calculate body mass index (BMI, weight [kilograms]/height [square meters]). Fixation pattern was assessed in very young patients, but their data were omitted from the final statistical analysis, because the comparison of initial and final visual acuities was impossible. The preferred method of visual field testing was the Humphrey SITA-FAST 30–2 visual field. Since this method is not feasible in very young children, they were tested by Goldmann perimetry or by confrontation, the latter performed by an experienced neuro-ophthalmologist (A.K.). In order to grade visual acuity, visual field, and degree of optic disc edema, a uniform system was developed and implemented for this study (Tables 1 and 2). The patients were followed-up in the neuro-ophthalmology unit and detailed information on visual acuity, visual field, and episodes of recurrence was recorded. During follow-up, patients and parents were carefully questioned about disease symptoms indicative of recurrence, as defined in the next paragraph.

Table 1 Visual acuity grading
Full size table
Table 2 Visual field grading
Full size table

All the pediatric IIH patients were initially treated with oral acetazolamide 15 mg/kg/day divided into three daily dosages, and some received additional sequential or combined treatment with topiramate. Treatment was augmented with steroids in cases where there was evidence of severe damage to the optic nerve. Patients with deterioration of visual function or intractable headache despite maximum treatment regimen were referred for surgical procedures (e.g., lumbo-peritoneal shunting procedures). Remission was defined as a symptom-free period of at least 4 weeks without evidence of optic disc edema and without concurrent medical treatment. An episode of recurrence was defined as a new post remission onset of disease symptoms or fundus findings of swollen discs occurring during free of treatment period. The disease symptoms included headaches, transient visual obscurations, and tinnitus. Neuroimaging and lumbar puncture were repeated in patients with recurrent episodes at clinician's discretion and according to the clinical setting.

Given that children aged 10 or less were very likely to be prepubertal, the study participants were divided into two age groups: 10 years or less and older than 10 years.

Statistical analyses

The statistical analysis was performed using SPSS software version 15.0 (SPSS Inc, Chicago, IL, USA). Each patient's visual acuity and visual field scores were averaged for statistical analysis. Visual acuity and visual fields were analyzed using Wilcoxon's non-parametric signed-rank test. Analysis of the incidence of IIH recurrence was performed using the Kaplan–Meier method. Cox regression analysis was used to calculate the correlation between duration of acetazolamide treatment and number of recurrences. Other continuous variables were analyzed using the two-tailed Student's t test.

Results

Patients

In total, 100 pediatric patients (age <18 years) were originally examined at our ophthalmology clinic between 1995 and 2009. We excluded four patients with insufficient data in their medical records, two patients with cells in their lumbar puncture, one patient with idiopathic hydrocephalus, one patient with venous sinus thrombosis, one patient with an arachnoid cyst, and one patient with an anaplastic astrocytoma. The remaining 90 patients comprised our study population. The mean age at onset of disease was 10 ± 5 years (mean ± standard deviation). Fifty of the children were ≤10 years of age at onset and forty were >10 years of age at onset. There were 45 males and 45 females (1:1 ratio). In the younger age group (≤10 years of age at onset), there were 29 males and 21 females. The mean follow-up period was 30.65 ± 27.47 months (range 1.15–172.6 months, median 25.18 months), and 27 (~33%) patients completed at least 3 years of follow-up. Fourteen patients had a follow-up period shorter than 6 months. Eight patients had only one visit. A total of six patients had been treated prior to presentation with medications suspected of being causative agents: three were treated with minocycline, two with isotretinoin, and one patient was treated with doxycycline.

All 90 children underwent cerebral imaging to rule out an intracranial space occupying lesion and sinus vein thrombosis. Computed tomography (CT) was performed in 50 patients. CT venography was performed in ten patients. Seventy-six patients underwent magnetic resonance imaging (MRI) and 47 underwent magnetic resonance venography. Sixty-five out of 90 patients underwent at least two imaging modalities. Table 3 includes major study parameters.

Table 3 Patient characteristics and main outcome measures
Full size table

Body mass index

Data on BMI at first visit was available for 53 patients. The mean BMI was 23.75 ± 6.89 kg/m2 (range 12.73–40.50). Nine patients had a BMI between 25 and 30, and 11 patients had a BMI over 30. Since BMI may not be valid in young children, BMI calculated for only those above age 10 years yielded a mean of 26.04 ± 6.3 with a similar distribution in the BMI groups above 25.

No correlation was found between BMI and visual acuity at the beginning of follow-up or at its end (p = 0.51, p = 0.054, respectively), nor did it correlate with visual field (p = 0.567, p = 0.224, respectively). A sub-analysis of the age group above 10 years yielded insignificant results.

The number of recurrences did not correlate with BMI (p = 0.7). This was also observed in the subgroup aged 10 years or above.

Visual findings

Forty two of 61 patients (~70%) had a normal visual field at presentation. One of the patients presented with optic atrophy in one eye, and three patients had macular involvement at onset. Fifty three out of 73 (~73%) patients had headache at onset, while 20 (~27%) were asymptomatic and IIH was detected incidentally. Seven patients complained of tinnitus, 18 patients had VIth nerve palsy, and 14 patients had transient visual obscurations.

Fundus examination findings

In order to describe the degree of optic disc edema, we used a scale, in which a normal optic disc was graded 0, mild edema was graded 1–2 according to its severity, fully developed edema was graded 3–4 with the latter representing full-blown optic disc edema. Chronic edema received the grade of 5, and optic atrophy was graded 6. Indeed, there are methodological flaws encountered when performing statistical analysis on these parameters, which by nature, are not continuous. Yet, with the lack of other alternatives, it does provide precise data, at least on the two extremities of the spectrum: normal discs versus chronic edematous discs or optic atrophy.

At presentation, the mean score of optic disc edema for the right eyes was 2.04 (±1.12) and for the left eyes 1.99 (±1.13). By the end of follow-up, the mean score for the right eyes was 1.18 (±1.4) and for the left eyes 1.25 (±1.52). This was highly significant in both eyes (p < 0.001). All patients had IIH with papilledema. Two out of 77 (2.6%) patients had an optic disc appearance of chronic disc edema at the end of follow-up. A total of seven eyes had optic atrophy at the end of follow-up (seven patients, three right eyes).

Lumbar puncture

Fifty five of 84 (~65%) patients underwent only one lumbar puncture. The normal upper opening pressure for adults is 25 cm H2O, but the maximal normal value in pediatric population has not been established. The highest opening pressure was between 20 and 24 cm H2O in 9 (11%) patients, between 25 and 39 cm H2O in 40 (48%) patients and exceeded 40 cm H2O in 35 (42%) patients. Among the children with opening pressure lower than 25 cm H2O, only one patient was older than 10 years of age. Six patients (6.67%) in this series had refractory disease (either persistent headaches or severe visual field loss in spite of maximal therapy) that necessitated lumbo-peritoneal shunting. None of the patients underwent ventriculo-peritoneal shunting or optic nerve sheath fenestration.

Outcome measures

A total of 67 out of 90 (74.4%) patients did not have any recurrence during the follow-up period. Twenty three of 90 patients remained active throughout follow-up. Thirteen patients had a single recurrent episode, 5 patients had two episodes and only 1 patient had three recurrent episodes. Information on first remission was available for 46 out of 90 patients. The mean duration between first visit and first remission was 11.78 ± 14.12 months (median 8.1 months). Assessment by a Kaplan–Meier survival analysis (Fig. 1) revealed that the overall risk of recurrence was 23.7%. The risk of recurrence was 9.3% ± 3.6% during the first year, then rose to 20% ± 5.2 after the second year and remained stable during the third year. A plateau was reached around 18 months of follow-up. There was no significant difference in recurrence rates between the younger (≤10 years) and older (>10 years) age groups (p = 0.6).

Fig. 1
figure 1

Incidence of idiopathic intracranial hypertension recurrence in the pediatric population (Kaplan–Meier survival analysis)

Full size image

Sixty patients were treated medically with acetazolamide alone (6 of these patients underwent lumbo-peritoneal shunting procedure). The mean duration of acetazolamide treatment was 12.99 ± 12.15 months. No correlation was found between the duration of treatment and the number of recurrent episodes (p = 0.25), and longer periods of treatment were not predictors of multiple recurrences. Cox regression analysis was performed to overcome the uneven follow-up time in different patients receiving acetazolamide treatment. Other medical treatment modalities were employed as well: 13 patients were treated with steroids, 8 patients were treated with topiramate, and 12 patients needed a combination of two drugs or more to control disease activity.

Patients with an initial score of central–steady–maintain, in whom the quantification of initial visual acuity status was not possible, were omitted from the evaluation of change in the visual acuity score between the first and last follow-up visits. Therefore, only 66 patients were included in the visual acuity analysis. The mean visual acuity score improved from 4.7 ± 0.62 to 4.87 ± 0.44, p = 0.003, representing an excellent outcome. The mean visual acuity improved in both age groups. While in the younger group, there was only a trend towards improvement (p = 0.06), it was significant in the older one (p = 0.009). However, patients whose visual acuity was severely compromised at presentation had no improvement in visual acuity at the end of follow-up.

Data on visual field scores both at onset and final follow-up were available for 48 of the 90 patients. The data included computed perimetries as well as Goldmann perimetry, and confrontational visual field assessment (the latter being relevant for six patients alone at presentation). The mean visual field score improved from 3.41 ± 0.8 to 3.52 ± 0.75, but this change did not reach a level of significance (p = 0.21). There was also no significant difference among the younger children (p = 0.751) nor the older ones (p = 0.7), although it is reasonable to assume that the initial visual field assessment did not worsen. We did not find any correlation between the risk of recurrence and the highest measurement of CSF opening pressure (p = 0.557).

Discussion

This study includes a total of 90 pediatric patients, with about half of them having been diagnosed before the age of 10 years. This makes this series one of the largest published. The follow-up period was relatively long, and the examinations performed mainly by one neuro-ophthalmologist (A.K.).

IIH is rare among children, and previous studies have suggested differences between the adult and the pediatric forms of the disease. Unlike the adult population, in which there is a strong female predominance [11], the ratio in children may be closer to 1:1, only to become more common in females among pubertal children and adolescents [9, 12]. Moreover, while the syndrome is rarely present in very young children [13], the incidence rises in adolescents [9, 14, 15]. One study has suggested puberty as being a risk factor for an unfavorable outcome [16]; however, some authors have suggested that the visual outcome in children may be better than that for adults [17, 18], and that spontaneous remission after diagnostic lumbar puncture may be more common [18]. The diagnosis of IIH in infants and young children is further complicated by atypical clinical presentations, such as irritability and somnolence [18], and by the obvious difficulties in obtaining reliable results of visual acuity and visual field testing, especially in very young children. The diagnostic CSF opening pressure is also a focus of debate, and the lack of consensus on what defines the values within a normal range for children complicates the diagnosis of IIH even more. A major review of the disorder suggested that the upper limit for normal CSF opening pressure is up to 18 cm H2O in children younger than 8 years, compared to up to 25 cm H2O for adults [19]. A recent publication [20] suggested that the maximal normal CSF opening pressure was 28 cm H2O for children. Another major review has suggested an opening pressure of 20 cm H2O to be diagnostic of IIH in children 8 years of age or younger [21].

The results of a retrospective study of adult IIH patients with a mean follow-up of 6.2 years [22], revealed recurrence rates of almost 40%. Recurrence did not occur until after the first 10 months from the time of diagnosis, and none occurred while the patients were being treated with acetazolamide. One study of IIH in a pediatric population described the risk of recurrence as being 20% [23]. In their review of the literature, Rangwala and Liu reported that the risk of recurrence ranged between 6% and 22% [19]. These reports, however, included a small number of patients.

The results of our study demonstrated a fairly benign disease course in the pediatric population, with no significant difference between the prepubertal and pubertal patients. The overall male/female ratio was equal. In patients ≤10 years of age at onset, the ratio was close to 1:1 in agreement with the findings of previous studies [24]. The BMI in our pediatric group was not substantially abnormal, especially in the age group above 10 years, while in other studies, the BMI in children was found to be usually normal [9, 24]. The mean visual acuity of our entire cohort had significantly improved at the end of follow-up period. We used visual acuity measurements applied in clinical practice (i.e., Snellen), owing to the retrospective nature of this study. Stiebel-Kalish et al. [16] suggested that post-pubertal children have a better outcome, but the parameter of secondary sexual development signs had not been addressed in their study. We were unable to test their findings due to insufficient documentation of Tanner signs for secondary sexual development in our data. We believe that inaccuracies of visual function testing in very young children may mask a better outcome than that of older children.

Our results showed that the risk of IIH recurrence in the pediatric population was about 24%, in agreement with the findings of Tibussek et al. [23]. In our series, the recurrence rates peaked during the first year and a half after diagnosis, and then decreased with time, giving us reason to assume that there is a relatively limited likelihood of developing a recurrence after the 18 months from diagnosis. It may be argued that these outcomes are a result of a mean follow-up of about 2.5 years, but the yearly decrements in recurrence rates (Fig. 1.) suggest that additional follow-up would not have substantially altered these results.

In an earlier study on adult IIH, the risk of recurrence was about 40%, and the recurrences were characterized as being late, i.e., occurring at least 10 months after the initial presentation [22]. In our study, the overall risk of recurrence was 23.7%, the recurrences were most common during the 18 months from diagnosis, and the overall prognosis was favorable. Considering that the etiology of this disease is unknown, the wide differences in the rates of recurrence between the children and adults may imply differences in IIH etiology for the two groups.

The highest CSF pressures measured during lumbar puncture were not predictors of outcome. In fact, higher CSF opening pressures did not correlate well with the number of recurrences. Longer duration of oral acetazolamide treatment was also not predictive of a poor outcome.

Only six patients (6.67%) in this series had refractory disease and underwent lumbo-peritoneal shunting. This suggests that most patients did not follow an aggressive disease course.

In conclusion, the current study adds more important information to the little known about the natural history of IIH in the pediatric age group. The children in this study had a favorable visual outcome, with minimal permanent impairments in their visual fields and visual acuity. The dreaded outcome of optic atrophy was rare. Furthermore, recurrence occurring after the first 18 months from diagnosis was highly unusual. Based on these findings, it appears that the outcome of IIH among children is better than it is for adults.